Drop setUpClass in reward tester
#1895
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
… into drop_setup_in_test
kashif
approved these changes
Aug 5, 2024
qgallouedec
added a commit
to yumeng5/trl
that referenced
this pull request
Aug 18, 2024
yxliu-TAMU
pushed a commit
to mincheolseong/ECEN743-GRPO-Project-Proposal
that referenced
this pull request
Apr 20, 2025
* drop setUp class in reward tester * report to none * style --------- Co-authored-by: Quentin Gallouédec <quentin.gallouedec@huggingface.co>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Main change
I realized that the tests interfered with each other because they shared the same model object.
(For a bit of context: a test failed when it was named
my_test2, but not when it was namedmy_test, as the tests are run in alphabetical order).The solution is to reload the model each time, which since it's cached, isn't an overhead.
Should we propage this change everywhere?
Also
Do not report to wandb: when tests are not run in parallel, wandb instances interfere, and create weird errors, e.g. :
This problem doesn't occur in the CI, probably because they are run in parallel, but locally it does occur.